From: Oliver Upton <oliver.upton@linux.dev>
To: David Matlack <dmatlack@google.com>
Cc: Marc Zyngier <maz@kernel.org>, James Morse <james.morse@arm.com>,
Alexandru Elisei <alexandru.elisei@arm.com>,
Suzuki K Poulose <suzuki.poulose@arm.com>,
Catalin Marinas <catalin.marinas@arm.com>,
Will Deacon <will@kernel.org>,
Quentin Perret <qperret@google.com>,
Ricardo Koller <ricarkol@google.com>,
Reiji Watanabe <reijiw@google.com>,
Ben Gardon <bgardon@google.com>,
Paolo Bonzini <pbonzini@redhat.com>,
Gavin Shan <gshan@redhat.com>, Peter Xu <peterx@redhat.com>,
Sean Christopherson <seanjc@google.com>,
linux-arm-kernel@lists.infradead.org,
kvmarm@lists.cs.columbia.edu, kvm@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 06/14] KVM: arm64: Return next table from map callbacks
Date: Fri, 9 Sep 2022 10:38:23 +0100 [thread overview]
Message-ID: <YxsJj3ojGyhNw5Jn@google.com> (raw)
In-Reply-To: <YxkN7XmHiU3ddknR@google.com>
Hi David,
On Wed, Sep 07, 2022 at 02:32:29PM -0700, David Matlack wrote:
> On Tue, Aug 30, 2022 at 07:41:24PM +0000, Oliver Upton wrote:
> > The map walkers install new page tables during their traversal. Return
> > the newly-installed table PTE from the map callbacks to point the walker
> > at the new table w/o rereading the ptep.
> >
> > Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
> > ---
> > arch/arm64/kvm/hyp/pgtable.c | 9 +++++----
> > 1 file changed, 5 insertions(+), 4 deletions(-)
> >
> > diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c
> > index 331f6e3b2c20..f911509e6512 100644
> > --- a/arch/arm64/kvm/hyp/pgtable.c
> > +++ b/arch/arm64/kvm/hyp/pgtable.c
> > @@ -202,13 +202,12 @@ static inline int __kvm_pgtable_visit(struct kvm_pgtable_walk_data *data,
> > if (!table && (flags & KVM_PGTABLE_WALK_LEAF)) {
> > ret = kvm_pgtable_visitor_cb(data, addr, level, ptep, &pte,
> > KVM_PGTABLE_WALK_LEAF);
> > - pte = *ptep;
> > - table = kvm_pte_table(pte, level);
> > }
> >
> > if (ret)
> > goto out;
>
> Rather than passing a pointer to the local variable pte and requiring
> all downstream code to update it (and deal with dereferencing to read
> the old pte), wouldn't it be simpler to just re-read the PTE here?
Yeah, you're right. I had some odd rationalization about this, but
there's no need to force a walker to descend into the new table level as
it is wasted work if another thread unlinks it.
[...]
> >
> > + table = kvm_pte_table(pte, level);
> > if (!table) {
>
> nit: Technically there's no reason to set @table again. e.g. This could
> just be:
>
> if (!kvm_pte_table(pte, level)) {
Sure, I'll squish these lines together.
--
Thanks,
Oliver
WARNING: multiple messages have this Message-ID (diff)
From: Oliver Upton <oliver.upton@linux.dev>
To: David Matlack <dmatlack@google.com>
Cc: kvm@vger.kernel.org, Marc Zyngier <maz@kernel.org>,
linux-kernel@vger.kernel.org,
Catalin Marinas <catalin.marinas@arm.com>,
Ben Gardon <bgardon@google.com>,
Paolo Bonzini <pbonzini@redhat.com>,
Will Deacon <will@kernel.org>,
kvmarm@lists.cs.columbia.edu,
linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH 06/14] KVM: arm64: Return next table from map callbacks
Date: Fri, 9 Sep 2022 10:38:23 +0100 [thread overview]
Message-ID: <YxsJj3ojGyhNw5Jn@google.com> (raw)
In-Reply-To: <YxkN7XmHiU3ddknR@google.com>
Hi David,
On Wed, Sep 07, 2022 at 02:32:29PM -0700, David Matlack wrote:
> On Tue, Aug 30, 2022 at 07:41:24PM +0000, Oliver Upton wrote:
> > The map walkers install new page tables during their traversal. Return
> > the newly-installed table PTE from the map callbacks to point the walker
> > at the new table w/o rereading the ptep.
> >
> > Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
> > ---
> > arch/arm64/kvm/hyp/pgtable.c | 9 +++++----
> > 1 file changed, 5 insertions(+), 4 deletions(-)
> >
> > diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c
> > index 331f6e3b2c20..f911509e6512 100644
> > --- a/arch/arm64/kvm/hyp/pgtable.c
> > +++ b/arch/arm64/kvm/hyp/pgtable.c
> > @@ -202,13 +202,12 @@ static inline int __kvm_pgtable_visit(struct kvm_pgtable_walk_data *data,
> > if (!table && (flags & KVM_PGTABLE_WALK_LEAF)) {
> > ret = kvm_pgtable_visitor_cb(data, addr, level, ptep, &pte,
> > KVM_PGTABLE_WALK_LEAF);
> > - pte = *ptep;
> > - table = kvm_pte_table(pte, level);
> > }
> >
> > if (ret)
> > goto out;
>
> Rather than passing a pointer to the local variable pte and requiring
> all downstream code to update it (and deal with dereferencing to read
> the old pte), wouldn't it be simpler to just re-read the PTE here?
Yeah, you're right. I had some odd rationalization about this, but
there's no need to force a walker to descend into the new table level as
it is wasted work if another thread unlinks it.
[...]
> >
> > + table = kvm_pte_table(pte, level);
> > if (!table) {
>
> nit: Technically there's no reason to set @table again. e.g. This could
> just be:
>
> if (!kvm_pte_table(pte, level)) {
Sure, I'll squish these lines together.
--
Thanks,
Oliver
_______________________________________________
kvmarm mailing list
kvmarm@lists.cs.columbia.edu
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
WARNING: multiple messages have this Message-ID (diff)
From: Oliver Upton <oliver.upton@linux.dev>
To: David Matlack <dmatlack@google.com>
Cc: Marc Zyngier <maz@kernel.org>, James Morse <james.morse@arm.com>,
Alexandru Elisei <alexandru.elisei@arm.com>,
Suzuki K Poulose <suzuki.poulose@arm.com>,
Catalin Marinas <catalin.marinas@arm.com>,
Will Deacon <will@kernel.org>,
Quentin Perret <qperret@google.com>,
Ricardo Koller <ricarkol@google.com>,
Reiji Watanabe <reijiw@google.com>,
Ben Gardon <bgardon@google.com>,
Paolo Bonzini <pbonzini@redhat.com>,
Gavin Shan <gshan@redhat.com>, Peter Xu <peterx@redhat.com>,
Sean Christopherson <seanjc@google.com>,
linux-arm-kernel@lists.infradead.org,
kvmarm@lists.cs.columbia.edu, kvm@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 06/14] KVM: arm64: Return next table from map callbacks
Date: Fri, 9 Sep 2022 10:38:23 +0100 [thread overview]
Message-ID: <YxsJj3ojGyhNw5Jn@google.com> (raw)
In-Reply-To: <YxkN7XmHiU3ddknR@google.com>
Hi David,
On Wed, Sep 07, 2022 at 02:32:29PM -0700, David Matlack wrote:
> On Tue, Aug 30, 2022 at 07:41:24PM +0000, Oliver Upton wrote:
> > The map walkers install new page tables during their traversal. Return
> > the newly-installed table PTE from the map callbacks to point the walker
> > at the new table w/o rereading the ptep.
> >
> > Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
> > ---
> > arch/arm64/kvm/hyp/pgtable.c | 9 +++++----
> > 1 file changed, 5 insertions(+), 4 deletions(-)
> >
> > diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c
> > index 331f6e3b2c20..f911509e6512 100644
> > --- a/arch/arm64/kvm/hyp/pgtable.c
> > +++ b/arch/arm64/kvm/hyp/pgtable.c
> > @@ -202,13 +202,12 @@ static inline int __kvm_pgtable_visit(struct kvm_pgtable_walk_data *data,
> > if (!table && (flags & KVM_PGTABLE_WALK_LEAF)) {
> > ret = kvm_pgtable_visitor_cb(data, addr, level, ptep, &pte,
> > KVM_PGTABLE_WALK_LEAF);
> > - pte = *ptep;
> > - table = kvm_pte_table(pte, level);
> > }
> >
> > if (ret)
> > goto out;
>
> Rather than passing a pointer to the local variable pte and requiring
> all downstream code to update it (and deal with dereferencing to read
> the old pte), wouldn't it be simpler to just re-read the PTE here?
Yeah, you're right. I had some odd rationalization about this, but
there's no need to force a walker to descend into the new table level as
it is wasted work if another thread unlinks it.
[...]
> >
> > + table = kvm_pte_table(pte, level);
> > if (!table) {
>
> nit: Technically there's no reason to set @table again. e.g. This could
> just be:
>
> if (!kvm_pte_table(pte, level)) {
Sure, I'll squish these lines together.
--
Thanks,
Oliver
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
next prev parent reply other threads:[~2022-09-09 9:38 UTC|newest]
Thread overview: 96+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-08-30 19:41 [PATCH 00/14] KVM: arm64: Parallel stage-2 fault handling Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` [PATCH 01/14] KVM: arm64: Add a helper to tear down unlinked stage-2 subtrees Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` [PATCH 02/14] KVM: arm64: Tear down unlinked stage-2 subtree after break-before-make Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-09-06 14:35 ` Quentin Perret
2022-09-06 14:35 ` Quentin Perret
2022-09-06 14:35 ` Quentin Perret
2022-09-09 10:04 ` Oliver Upton
2022-09-09 10:04 ` Oliver Upton
2022-09-09 10:04 ` Oliver Upton
2022-09-07 20:57 ` David Matlack
2022-09-07 20:57 ` David Matlack
2022-09-07 20:57 ` David Matlack
2022-09-09 10:07 ` Oliver Upton
2022-09-09 10:07 ` Oliver Upton
2022-09-09 10:07 ` Oliver Upton
2022-09-14 0:20 ` Ricardo Koller
2022-09-14 0:20 ` Ricardo Koller
2022-09-14 0:20 ` Ricardo Koller
2022-10-10 3:58 ` Oliver Upton
2022-10-10 3:58 ` Oliver Upton
2022-10-10 3:58 ` Oliver Upton
2022-08-30 19:41 ` [PATCH 03/14] KVM: arm64: Directly read owner id field in stage2_pte_is_counted() Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` [PATCH 04/14] KVM: arm64: Read the PTE once per visit Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` [PATCH 05/14] KVM: arm64: Split init and set for table PTE Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` [PATCH 06/14] KVM: arm64: Return next table from map callbacks Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-09-07 21:32 ` David Matlack
2022-09-07 21:32 ` David Matlack
2022-09-07 21:32 ` David Matlack
2022-09-09 9:38 ` Oliver Upton [this message]
2022-09-09 9:38 ` Oliver Upton
2022-09-09 9:38 ` Oliver Upton
2022-08-30 19:41 ` [PATCH 07/14] KVM: arm64: Document behavior of pgtable visitor callback Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` [PATCH 08/14] KVM: arm64: Protect page table traversal with RCU Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-09-07 21:47 ` David Matlack
2022-09-07 21:47 ` David Matlack
2022-09-07 21:47 ` David Matlack
2022-09-09 9:55 ` Oliver Upton
2022-09-09 9:55 ` Oliver Upton
2022-09-09 9:55 ` Oliver Upton
2022-08-30 19:41 ` [PATCH 09/14] KVM: arm64: Free removed stage-2 tables in RCU callback Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-08-30 19:41 ` Oliver Upton
2022-09-07 22:00 ` David Matlack
2022-09-07 22:00 ` David Matlack
2022-09-07 22:00 ` David Matlack
2022-09-08 16:40 ` David Matlack
2022-09-08 16:40 ` David Matlack
2022-09-08 16:40 ` David Matlack
2022-09-14 0:49 ` Ricardo Koller
2022-09-14 0:49 ` Ricardo Koller
2022-09-14 0:49 ` Ricardo Koller
2022-08-30 19:50 ` [PATCH 10/14] KVM: arm64: Atomically update stage 2 leaf attributes in parallel walks Oliver Upton
2022-08-30 19:50 ` Oliver Upton
2022-08-30 19:50 ` Oliver Upton
2022-08-30 19:51 ` [PATCH 11/14] KVM: arm64: Make changes block->table to leaf PTEs parallel-aware Oliver Upton
2022-08-30 19:51 ` Oliver Upton
2022-08-30 19:51 ` Oliver Upton
2022-09-14 0:51 ` Ricardo Koller
2022-09-14 0:51 ` Ricardo Koller
2022-09-14 0:51 ` Ricardo Koller
2022-09-14 0:53 ` Ricardo Koller
2022-09-14 0:53 ` Ricardo Koller
2022-09-14 0:53 ` Ricardo Koller
2022-08-30 19:51 ` [PATCH 12/14] KVM: arm64: Make leaf->leaf PTE changes parallel-aware Oliver Upton
2022-08-30 19:51 ` Oliver Upton
2022-08-30 19:51 ` Oliver Upton
2022-08-30 19:51 ` [PATCH 13/14] KVM: arm64: Make table->block " Oliver Upton
2022-08-30 19:51 ` Oliver Upton
2022-08-30 19:51 ` Oliver Upton
2022-08-30 19:52 ` [PATCH 14/14] KVM: arm64: Handle stage-2 faults in parallel Oliver Upton
2022-08-30 19:52 ` Oliver Upton
2022-08-30 19:52 ` Oliver Upton
2022-09-06 10:00 ` [PATCH 00/14] KVM: arm64: Parallel stage-2 fault handling Marc Zyngier
2022-09-06 10:00 ` Marc Zyngier
2022-09-06 10:00 ` Marc Zyngier
2022-09-09 10:01 ` Oliver Upton
2022-09-09 10:01 ` Oliver Upton
2022-09-09 10:01 ` Oliver Upton
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=YxsJj3ojGyhNw5Jn@google.com \
--to=oliver.upton@linux.dev \
--cc=alexandru.elisei@arm.com \
--cc=bgardon@google.com \
--cc=catalin.marinas@arm.com \
--cc=dmatlack@google.com \
--cc=gshan@redhat.com \
--cc=james.morse@arm.com \
--cc=kvm@vger.kernel.org \
--cc=kvmarm@lists.cs.columbia.edu \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=maz@kernel.org \
--cc=pbonzini@redhat.com \
--cc=peterx@redhat.com \
--cc=qperret@google.com \
--cc=reijiw@google.com \
--cc=ricarkol@google.com \
--cc=seanjc@google.com \
--cc=suzuki.poulose@arm.com \
--cc=will@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.